[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-2199":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":16,"stars30d":17,"stars90d":16,"forks30d":16,"starsTrendScore":16,"compositeScore":18,"rankGlobal":10,"rankLanguage":10,"license":19,"archived":20,"fork":21,"defaultBranch":22,"hasWiki":21,"hasPages":21,"topics":23,"createdAt":10,"pushedAt":10,"updatedAt":24,"readmeContent":25,"aiSummary":26,"trendingCount":16,"starSnapshotCount":16,"syncStatus":27,"lastSyncTime":28,"discoverSource":29},2199,"Detectron","facebookresearch\u002FDetectron","facebookresearch","FAIR's research platform for object detection research, implementing popular algorithms like Mask R-CNN and RetinaNet.","",null,"Python",26375,5389,1,306,0,3,45,"Apache License 2.0",true,false,"main",[],"2026-06-12 02:00:38","**Detectron is deprecated. Please see [detectron2](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdetectron2), a ground-up rewrite of Detectron in PyTorch.**\n\n# Detectron\n\nDetectron is Facebook AI Research's software system that implements state-of-the-art object detection algorithms, including [Mask R-CNN](https:\u002F\u002Farxiv.org\u002Fabs\u002F1703.06870). It is written in Python and powered by the [Caffe2](https:\u002F\u002Fgithub.com\u002Fcaffe2\u002Fcaffe2) deep learning framework.\n\nAt FAIR, Detectron has enabled numerous research projects, including: [Feature Pyramid Networks for Object Detection](https:\u002F\u002Farxiv.org\u002Fabs\u002F1612.03144), [Mask R-CNN](https:\u002F\u002Farxiv.org\u002Fabs\u002F1703.06870), [Detecting and Recognizing Human-Object Interactions](https:\u002F\u002Farxiv.org\u002Fabs\u002F1704.07333), [Focal Loss for Dense Object Detection](https:\u002F\u002Farxiv.org\u002Fabs\u002F1708.02002), [Non-local Neural Networks](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.07971), [Learning to Segment Every Thing](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.10370), [Data Distillation: Towards Omni-Supervised Learning](https:\u002F\u002Farxiv.org\u002Fabs\u002F1712.04440), [DensePose: Dense Human Pose Estimation In The Wild](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.00434), and [Group Normalization](https:\u002F\u002Farxiv.org\u002Fabs\u002F1803.08494).\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\"demo\u002Foutput\u002F33823288584_1d21cf0a26_k_example_output.jpg\" width=\"700px\" \u002F>\n  \u003Cp>Example Mask R-CNN output.\u003C\u002Fp>\n\u003C\u002Fdiv>\n\n## Introduction\n\nThe goal of Detectron is to provide a high-quality, high-performance\ncodebase for object detection *research*. It is designed to be flexible in order\nto support rapid implementation and evaluation of novel research. Detectron\nincludes implementations of the following object detection algorithms:\n\n- [Mask R-CNN](https:\u002F\u002Farxiv.org\u002Fabs\u002F1703.06870) -- *Marr Prize at ICCV 2017*\n- [RetinaNet](https:\u002F\u002Farxiv.org\u002Fabs\u002F1708.02002) -- *Best Student Paper Award at ICCV 2017*\n- [Faster R-CNN](https:\u002F\u002Farxiv.org\u002Fabs\u002F1506.01497)\n- [RPN](https:\u002F\u002Farxiv.org\u002Fabs\u002F1506.01497)\n- [Fast R-CNN](https:\u002F\u002Farxiv.org\u002Fabs\u002F1504.08083)\n- [R-FCN](https:\u002F\u002Farxiv.org\u002Fabs\u002F1605.06409)\n\nusing the following backbone network architectures:\n\n- [ResNeXt{50,101,152}](https:\u002F\u002Farxiv.org\u002Fabs\u002F1611.05431)\n- [ResNet{50,101,152}](https:\u002F\u002Farxiv.org\u002Fabs\u002F1512.03385)\n- [Feature Pyramid Networks](https:\u002F\u002Farxiv.org\u002Fabs\u002F1612.03144) (with ResNet\u002FResNeXt)\n- [VGG16](https:\u002F\u002Farxiv.org\u002Fabs\u002F1409.1556)\n\nAdditional backbone architectures may be easily implemented. For more details about these models, please see [References](#references) below.\n\n## Update\n\n- 4\u002F2018: Support Group Normalization - see [`GN\u002FREADME.md`](.\u002Fprojects\u002FGN\u002FREADME.md)\n\n## License\n\nDetectron is released under the [Apache 2.0 license](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdetectron\u002Fblob\u002Fmaster\u002FLICENSE). See the [NOTICE](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdetectron\u002Fblob\u002Fmaster\u002FNOTICE) file for additional details.\n\n## Citing Detectron\n\nIf you use Detectron in your research or wish to refer to the baseline results published in the [Model Zoo](MODEL_ZOO.md), please use the following BibTeX entry.\n\n```\n@misc{Detectron2018,\n  author =       {Ross Girshick and Ilija Radosavovic and Georgia Gkioxari and\n                  Piotr Doll\\'{a}r and Kaiming He},\n  title =        {Detectron},\n  howpublished = {\\url{https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdetectron}},\n  year =         {2018}\n}\n```\n\n## Model Zoo and Baselines\n\nWe provide a large set of baseline results and trained models available for download in the [Detectron Model Zoo](MODEL_ZOO.md).\n\n## Installation\n\nPlease find installation instructions for Caffe2 and Detectron in [`INSTALL.md`](INSTALL.md).\n\n## Quick Start: Using Detectron\n\nAfter installation, please see [`GETTING_STARTED.md`](GETTING_STARTED.md) for brief tutorials covering inference and training with Detectron.\n\n## Getting Help\n\nTo start, please check the [troubleshooting](INSTALL.md#troubleshooting) section of our installation instructions as well as our [FAQ](FAQ.md). If you couldn't find help there, try searching our GitHub issues. We intend the issues page to be a forum in which the community collectively troubleshoots problems.\n\nIf bugs are found, **we appreciate pull requests** (including adding Q&A's to `FAQ.md` and improving our installation instructions and troubleshooting documents). Please see [CONTRIBUTING.md](CONTRIBUTING.md) for more information about contributing to Detectron.\n\n## References\n\n- [Data Distillation: Towards Omni-Supervised Learning](https:\u002F\u002Farxiv.org\u002Fabs\u002F1712.04440).\n  Ilija Radosavovic, Piotr Dollár, Ross Girshick, Georgia Gkioxari, and Kaiming He.\n  Tech report, arXiv, Dec. 2017.\n- [Learning to Segment Every Thing](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.10370).\n  Ronghang Hu, Piotr Dollár, Kaiming He, Trevor Darrell, and Ross Girshick.\n  Tech report, arXiv, Nov. 2017.\n- [Non-Local Neural Networks](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.07971).\n  Xiaolong Wang, Ross Girshick, Abhinav Gupta, and Kaiming He.\n  Tech report, arXiv, Nov. 2017.\n- [Mask R-CNN](https:\u002F\u002Farxiv.org\u002Fabs\u002F1703.06870).\n  Kaiming He, Georgia Gkioxari, Piotr Dollár, and Ross Girshick.\n  IEEE International Conference on Computer Vision (ICCV), 2017.\n- [Focal Loss for Dense Object Detection](https:\u002F\u002Farxiv.org\u002Fabs\u002F1708.02002).\n  Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He, and Piotr Dollár.\n  IEEE International Conference on Computer Vision (ICCV), 2017.\n- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.02677).\n  Priya Goyal, Piotr Dollár, Ross Girshick, Pieter Noordhuis, Lukasz Wesolowski, Aapo Kyrola, Andrew Tulloch, Yangqing Jia, and Kaiming He.\n  Tech report, arXiv, June 2017.\n- [Detecting and Recognizing Human-Object Interactions](https:\u002F\u002Farxiv.org\u002Fabs\u002F1704.07333).\n  Georgia Gkioxari, Ross Girshick, Piotr Dollár, and Kaiming He.\n  Tech report, arXiv, Apr. 2017.\n- [Feature Pyramid Networks for Object Detection](https:\u002F\u002Farxiv.org\u002Fabs\u002F1612.03144).\n  Tsung-Yi Lin, Piotr Dollár, Ross Girshick, Kaiming He, Bharath Hariharan, and Serge Belongie.\n  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.\n- [Aggregated Residual Transformations for Deep Neural Networks](https:\u002F\u002Farxiv.org\u002Fabs\u002F1611.05431).\n  Saining Xie, Ross Girshick, Piotr Dollár, Zhuowen Tu, and Kaiming He.\n  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.\n- [R-FCN: Object Detection via Region-based Fully Convolutional Networks](http:\u002F\u002Farxiv.org\u002Fabs\u002F1605.06409).\n  Jifeng Dai, Yi Li, Kaiming He, and Jian Sun.\n  Conference on Neural Information Processing Systems (NIPS), 2016.\n- [Deep Residual Learning for Image Recognition](http:\u002F\u002Farxiv.org\u002Fabs\u002F1512.03385).\n  Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun.\n  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.\n- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](http:\u002F\u002Farxiv.org\u002Fabs\u002F1506.01497)\n  Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun.\n  Conference on Neural Information Processing Systems (NIPS), 2015.\n- [Fast R-CNN](http:\u002F\u002Farxiv.org\u002Fabs\u002F1504.08083).\n  Ross Girshick.\n  IEEE International Conference on Computer Vision (ICCV), 2015.\n","Detectron是Facebook AI Research开发的一个面向目标检测研究的软件系统，实现了包括Mask R-CNN和RetinaNet在内的多种先进算法。该项目基于Python语言，并利用Caffe2深度学习框架提供支持，其核心功能涵盖了从快速研发到评估新研究方法的全过程，适用于需要高精度与高性能图像识别的应用场景，如自动驾驶、安防监控等。尽管Detectron已不再维护并推荐使用基于PyTorch重写的detectron2版本，但其在推动计算机视觉领域尤其是物体检测方面仍具有重要参考价值。",2,"2026-06-11 02:48:48","top_language"]